This message was deleted.

Just found out that the nodes with no backups are probably giving up because of malformed cron syntax.